By: Visagen Naidoo - Mindworx Consulting

South Africa has a youth unemployment crisis. Everyone knows that. What few people are talking about is that we also have a programme design crisis. For years, our approach to early-career development has been built on a simple assumption: get young people into the workplace as quickly as possible and experience will do the rest. That assumption no longer holds.

AI is changing something fundamental about the labour market. It is quietly removing the first rung of the career ladder.

The World Economic Forum’s Future of Jobs Report 2025 paints a stark picture. Employers are increasingly hiring for AI skills while simultaneously reducing roles where technology can automate routine tasks. Businesses need more digital talent than ever before, but the entry-level work that once helped people build that talent is shrinking.

South Africa sits at the centre of this contradiction. We have one of the highest youth unemployment rates in the world while simultaneously facing a growing shortage of digital, IT and AI capability. Organisations cannot find the skills they need, yet millions of young people cannot find a way into the job market.

Those are not separate problems and unless we redesign how young people enter the workforce, both will get worse. The solution is not how many young people we can move through programmes but whether those programmes are designed for the world of work that now exists.

For too long, we have mistaken exposure for development. We take learners, graduates and interns, place them into complex environments and assume capability will emerge through proximity. We expect them to absorb business language, workplace dynamics and technical expectations while simultaneously proving their value.

Some thrive. Many simply survive. Across the market, we see the same pattern repeat itself. Bright, capable young people lose confidence within months, not because they lack potential, but because they have been dropped into systems they were never properly prepared to navigate. The result is predictable. They become quieter. They ask fewer questions. They contribute less. They remain visible, but they are not developing.

If AI is changing the nature of entry-level work, then we have to change the way we prepare people for it. That means confidence, context and capability can no longer be treated as things young people acquire accidentally along the way. They have to be designed intentionally into the experience. This is exactly what we have learned through our Mindworx Digital Pods.

Three years ago, our ambition was to place 5,000 learners and positively impact 25,000 lives by 2028. Today, that ambition has doubled to 10,000 learners and 50,000 lives by 2030. To be honest, we are behind. But that is exactly why I am optimistic. Because scaling has taught us something uncomfortable: bigger programmes are not the answer. Better-designed programmes are.

The pressure on learnership and graduate programmes is intensifying from every direction. AI is removing the bottom rung of the workforce. Clients want productive learners faster. Government wants numbers. Funders want impact.

At the same time, the regulatory environment is becoming more complex. The QCTO’s overhaul of occupational qualifications is reshaping what counts as a qualification, what is accredited and the timelines organisations now have to work within. It is a necessary evolution, but it has introduced a layer of complexity and uncertainty that rarely appears in client briefs or board reports.

The instinctive response to all of this pressure is to compress the process even further. Push learners into projects earlier. Increase exposure. Shorten the runway between training and delivery. The reality now is that speed without sequencing simply accelerates failure.

What our Digital Pods has taught us is that scale is a design problem before it is a logistics problem. Impact at scale only happens when capability is built deliberately, not assumed.

At smaller volumes, weak programme design can be rescued by exceptional mentors, patient clients and heroic programme managers. At scale, those interventions disappear. The system itself has to work and that changes everything.

It changes what we measure. Readiness matters more than exposure hours. Contribution matters more than attendance. Confidence matters as much as technical competence.

It also changes what businesses need to provide for entry-level success. Young people do not simply need placements. They need environments deliberately designed for development. And that changes the question entirely.

The question is no longer how quickly we can reach 10,000 learners. It is whether the 10,000th learner receives the same quality of development as the 100th. Get the design wrong, and 10,000 simply becomes a vanity metric.

Every organisation trying to execute digital transformation is now competing for the same scarce capability. Every CIO is under pressure to deliver more with constrained budgets and limited talent pipelines. South Africa will not solve that problem by waiting for universities, government or another generation of graduates to eventually catch up. We have to build a better on-ramp into digital work now.

Because if we continue building programmes for an economy that no longer exists, the consequences will be profound. South Africa does not need more programmes, it needs better-designed ones. The first rung of the career ladder is disappearing. The urgency now is to build a new one.